基于RBF神经网络的微机械陀螺仪温度控制系统

为了避免微机械陀螺仪受温度影响产生较大波动,设计了基于RBF神经网络的微机械陀螺仪温度控制系统。首先,构建了微机械陀螺仪温度控制装置的数学模型;其次,鉴于控制模型参数会随温度改变而发生变化,提出了基于RBF神经网络的自适应PID控制算法,利用径向基函数的局部逼近能力调节微机械陀螺仪环境温度。结果表明,设计方案基于RBF神经网络,可实时整定PID参数,且能够准确控制微机械陀螺仪的温度。与传统PID控制相比,基于RBF神经网络的自适应PID控制算法具有更快的调节速度和较强的鲁棒性,实现了微机械陀螺仪温度的无静差控制,降低了模型参数变化对控制效果的影响,满足温度控制系统的要求。

Temperature control system of micro-mechanical gyroscope based on RBF neural network

In order to avoid large fluctuation of the micro-mechanical gyroscope due to temperature variation, a micro-mechanical gyroscope temperature control system based on RBF neural network is established. First of all, a mathematical model of micro-mechanical gyroscope temperature control device is constructed. Secondly, considering that the parameter of control model is changing with temperature, an adaptive PID control algorithm based on RBF neural network is proposed. The local approximation ability of radial basis function can adjust ambient temperature of the micro-mechanical gyroscope. The method uses RBF neural network to adjust PID parameters in real time and control the temperature of micro-mechanical gyroscope accurately. Compared with traditional PID control, the adaptive PID control algorithm based on RBF neural network has faster adjustment speed and stronger robustness, and can realize static error free control of the temperature of micro-mechanical gyroscope and reduce the variation of model parameters. The designed control system satisfies the requirements of temperature control.